Does AM2 Reduce the Impact of L2 Cache Size?

We've already seen that Socket-AM2 doesn't really impact performance except for in games, but does the higher bandwidth memory controller reduce the impact of AMD's 1MB L2 cache parts compared to its 512KB L2 cache offerings? 

 Benchmark - Athlon 64 X2 2.0GHz Socket-939 (1MB vs. 512KB Advantage)  Socket-AM2 (1MB vs. 512KB Advantage)
Cinebench 9.5 Multi-Core Rendering Test 0.2% 0%
3dsmax 7 0.3% 0.6%
Adobe Photoshop CS2 2.7% 2.5%
DivX 6.1 0% 0%
WME9 0% 1%
Quicktime 7.0.4 (H.264) 0.9% 1.3%
iTunes (MP3) 0% 0%
Quake 4 - 10x7 (SMP) 4.8% 3.5%
Oblivion - 10x7 7.5% 3.3%
F.E.A.R. - 10x7 8.6% 6.2%


In the application benchmarks there isn't really a difference in how performance scales with cache size between the two platforms, but looking at the games there is indication of a pattern that is developing.

In Quake 4, Oblivion and F.E.A.R. the 1MB L2 cache seems to make slightly more of a difference on the Socket-939 platform than on the Socket-AM2 platform.  While the 1MB cache offers a 4.8%, 7.5% and 8.6% performance advantage in those three games on the Socket-939 platform, on AM2 the advantage is cut down to 3.5%, 3.3% and 6.2% respectively.  The explanation being that with a lower latency memory controller and more available memory bandwidth, the benefits of a larger cache are reduced on Socket-AM2. 

However the differences in performance scaling that we're seeing here are small enough that once you take into account the amount of variation you can see between runs, it's not really worth concluding anything concrete based on this data.  What we do see here is a trend of the 1MB L2 cache parts doing less on Socket-AM2 than on Socket-939 (another way of looking at it is that the 512KB are doing better on AM2 than they did on 939), but the margins are small enough that we can't really say for sure what is causing the trend.

Once again, the trend only seems to impact games, as the other application tests we've run appear to be basically unaffected. 

The Question on Everyone's Mind: Is AM2 Faster? How Does the New 4000+ Stack Up?


View All Comments

  • mino - Tuesday, May 23, 2006 - link

    2 options:
    1) Quad-core K8 (on 65nm)
    2) High-speed 65nm DC's(improbable)

    BTW what I understand FX-64 is on the way in a few months(july-august). Seems rev. F cores could handle 3.0, just 125W TDP may be the issue.
  • peternelson - Tuesday, May 23, 2006 - link

    Well, I don't think it's a QUAD CORE K8 (aka "Deerhound") because that is not due until late 2007.

    And dualcore K8L is not until 1H/2007.

    We need to choose something happening THIS year.
  • mino - Wednesday, May 24, 2006 - link

    Actually AMD can made Quad-core CPU's even on 90nm if they need to. The core will be huge, yields poor but IMHO 2.2G Quad at 90nm is possible within 125W TDP.

    Also AFAIK AMD has delayed 65nm at least for a quarter intentionally since what they need now is capacity on 90nm. They could not afford any (even short-term) production reduction at this moment. Provided in 3Q/06 FAB 36 is up and running at 10k starts the could afford to dedicate some of them for some high-end opterons and FX's.
  • Furen - Wednesday, May 24, 2006 - link">

    Look at that and tell me how you can possibly fit twice that (90nm dual-core) in one package. Dual-core CPUs are huge to begin with, doubling the number of cores would probably require a pretty big drop in L2 sizes (think 256KB per core...). AMD still is production limited and designing a quad-core chip without going to 65nm would pretty much doom it to being a VERY low-volume part. Heck, Intel's Conroe is huge as well, it's just on a smaller process (the 160sq. mm die size would correspond to something like 300+sq. mm on the 90nm process).
  • jones377 - Tuesday, May 23, 2006 - link

    It's called Socket F Reply
  • Griswold - Tuesday, May 23, 2006 - link

    I dont think so. Socket F isnt really a "secret" nor a stopgap solution. Reply
  • peternelson - Tuesday, May 23, 2006 - link

    I think we hear more about socket F in June and it launches July.

    But that's not what this is alluding to.

    There was an announcement of a roadmap change from Q1/2007 to DECEMBER 2006.

    If I remember right it was two AM2 processors on 65 nanometre process.
  • jones377 - Tuesday, May 23, 2006 - link

    Perhaps not, but it's coming out at around that timeframe. Anything else and we would have gotten wind of it long ago. Reply
  • mlittl3 - Tuesday, May 23, 2006 - link

    A couple of things before I give my guess about the stopgap solution...

    1) K8L as state above WILL HAVE microarchitectural improvements. This has been all over the internet.

    2) AMD's processor pricing page states that the X2 5000+ and FX-62 will be available for both 939 and AM2. I don't know if they messed up but if not, it looks like 939 users can upgrade yet again.

    Okay, here's my guess for the stopgap solution...drum roll...L3 cache. I think AMD will release a 2.8 revised FX-62 with L3 cache or an ahead of schedule 3.0 GHz FX-64 with L3 cache. Just my guess.
  • AllYourBaseAreBelong2Us - Tuesday, May 23, 2006 - link

    The stopgap solution is the 65nm process that will allow AMD to ramp up the speed a bit more and get better TDP ratings.

Log in

Don't have an account? Sign up now